An updated version of this report can be found  here.
(April 16, 2002)

Banking on multinationals

Increased competition from large foreign lenders threatens domestic banks, raises financial instability

by Christian Weller

When the International Monetary Fund (IMF) and World Bank convene in Washington, D.C. next week to discuss their futures, one thing is certain – the recently released Meltzer report, which made many proposals for reshaping these institutions, will figure prominently on the agenda.

The Meltzer commission proposes that the IMF be transformed into a lender-of-last-resort, providing short-term emergency loans to pre-qualified countries. To pre-qualify for such loans, countries would have to develop their own financial markets first by adequately capitalizing their commercial banks and then by allowing the free entry and operation of foreign banks. While adequate capitalization is an uncontroversial proposal, recommendations for increased international financial competition deserve careful scrutiny. In particular, the hope that increased competition will make domestic banks more efficient, and thus more stable, is not supported by the existing economic evidence. In fact, research shows that increased foreign competition lowers the supply of credit and possibly raises financial instability.¹

Why do banks go global?
Multinational banks (MNBs), by definition, are those that physically operate in more than one country. For instance, Citibank operates offices in more than 90 countries around the world. In contrast, international banks engage in cross-border operations and do not set up operations in other countries. A Bank of America loan to a bank in Poland is considered international banking.

MNBs have experienced rapid growth in the past few years (see Table 1). Access to new markets in Central and Eastern Europe and financial deregulation elsewhere has helped to accelerate the growth of MNB operations. The most rapid growth of MNB loans has been in Eastern Europe, thanks to the opening of the former East Bloc countries. Latin America has also experienced a rapid increase in MNB loans over the years, following financial deregulation and bank privatizations.

The impressive growth of MNB loans does not reflect equally fast-growing capital inflows. MNBs basically work like this – they collect local currency, combine that with funds borrowed from overseas, and then lend that money domestically. Hence, MNBs are small net importers of capital, otherwise operating in a manner similar to domestic banks. Yet, even though MNBs look like domestic banks, their entry and operations are subject to international trade agreements, such as the General Agreement on Trade in Services (GATS) at the World Trade Organization (WTO).

MNBs tend to expand their global operations for two reasons. First, MNBs follow their clients. Since MNBs operate in a wide array of countries and regions, multinational corporations (MNCs) become their natural clients. When MNCs establish new operations, MNBs often follow. Second, the concentration on a small number of large clients in combination with superior technology and know-how makes MNB operations very profitable. For instance, even in the midst of high inflation and economic turmoil in Brazil in 1983, MNBs operated very profitably, with Citibank earning 20% of its worldwide income in Brazil that year. In Korea, MNBs averaged 75-80% returns on equity, whereas domestic banks earned only an average of 15-20% between 1972 and 1979. In 1991, MNB profits in Korea rose by 37% compared to 12.9% for domestic banks, leaving MNBs with a share of 68% of all net bank earnings, up from 61% in 1990.

Less industrialized economies also have become more open to MNBs, often because of a need for international capital. MNBs are, after all, a form of capital inflow and hence help to finance current account deficits. Furthermore, countries that have undergone financial crises are often looking to stabilize their banks with the help of international investors, often by allowing MNB operations to quickly expand their operations. In Korea and Thailand, MNB loans dipped briefly after the financial crisis hit, but within a year recovered and even accelerated (see Figure 1). A similar trend was also observed in Mexico, where MNB loans more than doubled from $1.8 billion at the beginning of the peso crisis in December 1994 to $4.3 billion in December 1995, only to jump to $8.0 billion by December 1996 (Figure 1). The need to find investors for troubled banks coupled with the fact that, after a rapid devaluation, domestic assets can be bought at bargain-basement prices helps explain the increase in MNB loans after a crisis.

In their operations, MNBs focus on a select range of activities for a small circle of clients. MNBs tend to provide services that other banks are either less familiar with or do not offer, such as foreign currency loans, acceptances and guarantees related to international trade, or syndicated loans. As a result, MNB clients are usually MNCs or large domestic corporations engaged in international transactions. In addition, MNBs also provide services for high-income earners, or what is referred to as high net-worth individuals.

MNBs shrink supplies of external financing and destabilize host economies
Retail banking services, such as small checking and savings accounts, mortgages, or small business loans, are rarely emphasized by MNBs, despite their claims to the contrary. In 1995, Creditanstalt, an Austrian MNB that is especially active in Eastern Europe, promoted itself as one of the few MNBs that engaged in retail banking services. However, when asked what the minimum loan size for a Creditanstalt customer in Poland would be, the answer was $100,000. Given the wage differentials between Poland and the United States, this would be equivalent to a minimum loan of more than $500,000 in the United States, well outside the range of a mass market. It is quite clear that MNBs do not serve the majority of low- and middle-income households or even small- and medium-sized enterprises and start-ups.

Even though MNBs only serve a small, wealthy circle of clients, the logic behind greater MNB entry is that domestic banks will be forced to improve their operations in order to remain competitive. According to this logic, which also underlies the Meltzer report’s recommendations, the resulting increase in the efficiency of domestic banks should increase the available amount of credit and improve the stability of local banks.

In practice, however, research finds that increased competition from MNBs leads domestic banks to reduce their loan exposure (Weller 2000a, 2000b; Weller and Scher 1999). Prime examples of this connection between more MNBs and less credit can be found in the economies of Central and Eastern Europe. In these areas MNBs have quickly gained significant market shares, while the credit supply, especially to smaller companies, has been stagnant or declining. The same result also holds true for a broad sample of economies in other parts of the world.

The fact that domestic banks lower their credit exposure is driven by the fact that MNBs “cherry pick” the best customers, leaving domestic banks with borrowers of lesser quality. As a result, both the costs and credit risks at domestic banks increase, while good, low-cost customers move to MNBs.

This puts domestic ban
ks in a bind. Thanks to increased international competition, the domestic banks inevitably need more capital to fund new technologies, to train existing workers, and to hire new banking experts. This greater international competition also means that domestic banks lose lending opportunities to MNCs and large domestic corporations – the most secure stream of income from loans – thereby making it harder for domestic banks to get the capital necessary to compete with MNBs.

As a result of lower credit and risk exposure, the stability of the banking sector increases, but only superficially. Since the most stable banks are the ones that do not lend any money, reduced lending also increases the stability of local banks. But this appearance of stability is misleading – in reality, little has changed in that economy’s banking technology or expertise that would allow domestic banks to better evaluate credit risks, primarily because the domestic banks do not have access to the funds needed for these types of improvements.

If domestic banks were to increase lending, then the stability of the banking sector would decline. Domestic banks may raise their loans again in the future, because the outlook for the respective economies has become more optimistic. Similarly, competitive pressures have grown to a point where banks need to begin to compete for market shares, even though they do not have sufficient capital to lend more. Thus, banks that shouldn’t be lending more will. In the end, domestic banks may look more stable, but this is due to temporarily lower loan exposure, not efficiency gains from greater competition.

Another destabilizing impact from declining loans results from the spill-over into the non-financial sector. Businesses need bank credit for investments. If loans are reduced due to increased international competition, firms will not have enough funds for their investments. As companies become unable to invest in new technologies, their ability to compete will decline, thereby weakening the non-financial sector and eventually the financial sector.²

But aren’t MNBs better and, hence, more stable banks? The Meltzer report states that “[f]oreign banks that were long-term direct investors in Brazil did not run [when the crisis occurred]; they acted as safe haven for frightened residents. Banking stability reduced capital flight, thereby limiting currency depreciation and the crises.” So according to this example, once the crisis occured, the presence of MNBs had a stabilizing impact on the host economy.

However, there is some evidence that MNBs occasionally foster the speculative bubbles that eventually lead to these crises. For example, Japanese MNBs speculated in real estate in Thailand in the early 1990s in an attempt to recover the losses incurred in Japan in the late 1980s, directly contributing to the financial crisis in Thailand in 1997. Thus, there is no clear evidence that MNBs finance necessarily sounder projects than domestic banks. And if the actions of MNBs force less well capitalized domestic banks to engage in speculative operations, MNBs may actually be destabilizing a country’s entire financial system.

Proponents of MNB entry claim that domestic banks will eventually become more efficient, but this assumption is based on the erroneous, neo-classical notion that banking technology and know-how will somehow naturally diffuse through the market. MNBs have technology and banking expertise that is obviously superior to that of local banks – that is part of what makes them MNBs. As would be expected, MNBs are aware of their competitive advantage and guard their technology and knowledge closely. In fact, the World Bank organized a “twinning” arrangement between local banks and MNBs in Poland in the early 1990s, but the arrangement failed because neither side wanted to nurture its competitors. One of the top priorities for MNBs is to establish branch offices or wholly owned subsidiaries that share as little information as possible with other banks. Banks are eager to keep their superior services and technologies closely guarded because these are core sources of competitive advantage in highly competitive markets.

Implications for public policy
So if the key to the superior performance of MNBs lies in their better technology and know-how, to gain access to these advantages host countries could impose capital controls that would require MNBs to enter into partnerships with domestic banks. For instance, MNBs could be required to establish local subsidiaries, while at the same time be prevented from owning the majority in any domestic company. Similarly, foreigners could be prevented from owning real estate, thus requiring them to seek out local partners if they want to own their branches. Such joint ventures may prove difficult to operate, however, due to different management and banking cultures and the reluctance of MNBs to share key technologies and information.

There is no real substitute for the development of an efficient domestic banking system. Such development requires the creation of stable institutions that will allocate funds to the most desirable uses. Fostering the creation of a local banking sector requires public financial support, as the experience of several developed economies has shown. Once these stable institutions are established, public policy should be designed to ensure that funds are directed toward sustainable development – infrastructure investments, education programs, modern environmental and medical technologies – rather than speculation.

Endnotes
1. This report is largely based on the findings in Weller (2000a, 2000b, 1999) and Weller and Scher (1999).

2. Weller (2000c) finds that increasing MNB loans have led to declining real credit in Central and Eastern Europe. Furthermore, this decline has increased the chance of banking crises.

References
Weller, Christian. 2000a. Financial liberalisation, multinational banks and credit supply. International Review of Applied Economics. Vol. 14, No. 2.

Weller, Christian. 2000b, “Multinational Banks in Developing and Transition Economies.” Economic Policy Institute, Technical Paper No. 241. Washington, D.C.: EPI.

Weller, Christian. 2000c. “Multinational banks in transition economies.” Paper presented at the Eastern Economic Association’s annual meeting, Crystal City, Virginia, March 2000.

Weller, Christian. 1999a. “The connection between more multinational banks and less credit in transition economies.” Center for European Integration Studies, ZEI Working Paper B99-8. Bonn, Germany: University of Bonn.

Weller, Christian, and Scher, Mark. 1999. “Multinational banks and development finance.” Center for European Integration Studies, ZEI Working Paper B99-16. Bonn, Germany: University of Bonn.

An updated version of this report can be found here.
(April 16, 2002)